3-tier Client/Server

Architecture – Operation Tiers

The system is based on Three Tier architecture and it will be installed using the TCP/IP communication protocol. The system consists of three discernible tiers:

  • Database Server
  • Application Server
  • Client



First Tier – Database Server

It constitutes the main part of the system; the Database Server provides all the necessary operations, retrieving, updating and supporting data. Moreover, it provides all the essential mechanisms for Data Integrity.

Second Tier – Application Server

It constitutes the main part of the hardware where the most operations are executed apart from those related to Screen format, (final user interface).

More than one Application Servers may be installed in different machines thus fully utilizing every available computational power, ensuring remarkable response, credibility and expansion results.

By distributing the Application Servers on independent machines, the overall load of the system is reduced. In this case every Server can support a subset of the Remote Clients total number (e.g. the Application Server A will serve the customers of Branch A, while the Application Server B will serve the customers of Branch B).

Third Tier – Client

The third software tier constitutes the user interface with the system. At this tier, the User Screens management is performed, as well as the format of data displayed. The Client’s communication with the Application or the Application Servers is performed using each time only one data package. The result is the best possible response time between the Client and the Application Server, since these two tiers can operate on a single telecommunications line (Leased Line, Dialup, and Internet Connection), ensuring small response times for the whole system.The Three-Tier Client – Server architecture has been proved internationally as the most appropriate for networked installations contrary to the Two-Tier Client architecture, or Fat Server architecture.

The three-tier composition ensures:
 

  • Reducing network’s overload, due to vast data volumes transfer. For example, running a Query to retrieve some documents from a table with several thousand entries is performed in the Application Server, which transfers to the client only the result.
  • Running the Database Server on a different machine from the Application Server or the Application Servers. This way, it becomes possible to define the crucial sizing of corresponding machine’s performance independently, while at the same time, you may ensure unlimited expansion without any software reconstruction
  • Maximum flexibility in the selection of the database server, since the use of any machine with any operating system (Windows NT ή UNIX etc), is permitted, requiring only the communication capability through TCP/IP protocol. Thus, a next upgrade is possible concerning the database by changing / upgrading the machine, without influencing the rest of the system.